Improved high-performance liquid chromatographic assay for the determination of ethionamide in serum

Journal of Chromatography, 563 (1991) 412-475 Biomedical Applications Elsevier Science Publishers B.V.. Amsterdam CHROMBIO.

5658

Short Communication Improved high-performance liquid chromatographic assay for the determination of ethionamide in serum CHARLES

A. PELOQUIN*

Department of Medicine, National Jewish Center,for Immunology and Respiratory Medicine, 1400 Jackson Street, Denver, CO 80206 (U.S.A.)*

and School of Pharmacy, University of Colorado, Denver, CO 80262

(U.S.A.) and GORDON

T. JAMES

and EILEEN

MCCARTHY

Department of Medicine, National Jewish

Centerfor Immunology and Respiratory Medicine, 1400 Jackson

Street, Denver, CO 80206 (U.S.A.) (First received July lOth, 1990; revised manuscript

received September

6th, 1990)

ABSTRACT A solid-phase

extraction

prior to high-performance

(SPE) method

used. Serum constituents

were removed

Samples were evaporated

to dryness,

ible, with a recovery

was developed

liquid chromatography

from the column

reconstituted

of ETA of 64%, comparable

to simplify

of ethionamide

the preparation

(ETA).

Octadecyl

of human

serum

SPE columns

were

with water, and ETA was eluted with methanol.

in mobile phase, and assayed. to the more tedious liquid-liquid

The method extraction

is reproducmethod

for

ETA.

INTRODUCTION

Ethionamide (ETA, 2-ethylthioisonicotinamide) is a second-line agent with activity against various mycobacteria, including M. tuberculosis, M. leprae, and M. avium-intracellulare (MAI) [1,2]. A number of assay methods for the measurement of ethionamide in serum have been developed [3-61. The method of Shepard et al. [5] has proven to be particularly useful because of its sensitivity, specificity, and reproducibility. We describe a solid-phase extraction (SPE) method which simplifies the preparation of serum for assay. EXPERIMENTAL

Chemicals Ethionamide 0378-4347/91/$03.50

and prothionamide 0

(PTA, internal standard)

1991 Elsevier Science Publishers

B.V

were donated

by

SHORT COMMUNICATIONS

473

Medimpex Northamerica (New York, NY, U.S.A.). Stock solutions (200 pg/ml) were prepared fresh daily by dissolving the drugs in 0.02 M disodium phosphate buffer-acetonitrile (5050, v/v). Extraction procedure

Aliquots (200 ~1) of serum were pipetted into 100 mm x 13 mm borosilicate tubes. The PTA internal standard stock solution (4 ,ul) was added to each sample. Samples were vortex-mixed for 5 s. Octadecyl SPE columns (1 ml) (J. T. Baker, Phillipsburg, NJ, U.S.A.) were prepared with one column volume of methanol followed by one column volume of purified water (Milli-Q system, Millipore, Milford, MA, U.S.A.). The 204~~1 samples were pipetted onto the columns and drawn onto the matrix using low vacuum (2 mmHg). The columns were washed with 400 ~1 of purified water to remove plasma constituents. Excess water was removed with a 50-~1 rinse of lo:90 (v/v) water-methanol, followed by drying under vacuum (15 mmHg) for 15 min. The collection rack holding 1.8-ml polypropylene collection vials was then placed into the vacuum manifold. ETA and PTA were extracted with 1000 ~1 of methanol using low vacuum (2 mmHg). After all of the methanol was drawn onto the matrix, the pressure was increased to 10 mmHg for 10 s. The collection vials were withdrawn from the vacuum manifold, placed in a water bath (40”(Z), and dried under nitrogen (approximately 15 min). The samples were reconstituted with 200 ~1 of 0.02 M disodium phosphate buffer-acetonitrile (75:25) and vortex-mixed for 5 s. The entire contents of each collection vial was transferred to an autosampler vial prior to analysis. Liquid chromatography

Analyses were performed using a Waters (Milford, MA, U.S.A.) Model M6000A pump, a Spectra-Physics (San Jose, CA, U.S.A.) Model 8875 autosampler, a Waters 440 ultraviolet detector set at 254 nm, 0.01 a.u.f.s., and a Rainin (Woburn, MA, U.S.A.) Dynamax@ HPLC Method Manager Data Package installed on a Macintosh Plus computer (Apple Computers, Cupertino, CA, U.S.A.). A reversed-phase system was used consisting of an Alltech (Deerfield, IL, U.S.A.) 5-pm Hypersil ODS cartridge column (250 mm x 4.6 mm I.D.) preceded by an Alltech 5-pm Hypersil ODS cartridge guard (10 mm x 4.6 mm I.D.) column. The mobile phase was 0.02 M disodium phosphate buffer-acetonitrile (75:25), which was filtered (0.45 pm), degassed by sonication, and delivered at a flow-rate of 1.5 ml/min (0.138 kPa). Injections of 20 ~1 were made, with a run time of 12 min per sample. Standard curves

Standard curves were constructed to encompass the anticipated range of serum concentrations found in patients taking ethionamide. Blank serum was spiked with ethionamide to give concentrations of 0.2,0.5, 1.O, 5.0, 10.0, and 20.0

414

SHORT

COMMUNICATIONS

/Ig/ml. Aliquots (200 ,ul) of each standard were then prepared using the above extraction procedure. The line of best fit was determined using weighted linear regression analysis of the ETA/PTA peak-height ratio (v) veY.sU.s the theoretical concentration of ETA (x). Selectivity

The selectivity of the method available in the United States, acetaminophen, was evaluated. above drugs were injected into water-acetonitrile (50:50).]

with respect to all other antimycobacterial agents as well as common agents such as aspirin and Aqueous solutions containing 20-75 pg/ml of the the analytical system. [Rifampin was dissolved in

RESULTS

Analvtical procedure

A representative chromatogram of the extract of a serum sample obtained 2 h after ingestion of 500 mg ethionamide is shown in Fig. 1. The retention times of ETA and PTA are 4.28 and 7.24 min, respectively. Separation of ETA and PTA was complete, with a resolution factor (I?,) of 5.93. Concentrations of ETA were determined using the peak-height ratio of ETA to PTA. Linear regression analysis using a weight of l/y2 provided the best fit over the 2 log concentration range. The three-day validation produced slopes ranging from 0.426 to 0.486 and intercepts ranging from 0.002 to 0.041. None of the intercepts differed significantly from zero. The response of the detector was linear over the 2 log range of concentrations tested, with a practical limit of detection of 0.2 pg/ml ETA (peak height-to-noise ratio 2 5:l). Overall recovery of ETA, calculated by comparing peak heights of extracted serum samples with those of directly injected water samples, was 64%. Analysis of the data for overall 0.012 ,

1

0.010 -

F 8

0.008 -

c? !

0.0°6 -

g

0.004 -

2

I (I) 3

0.002 0-d

L

’ 0

2

4

i 6

8

IO

12

Time (min) Fig. 1. Chromatogram of an extract of serum from a patient = ETA; 2 = PTA (internal standard).

2 h after the ingestion

of 500 mg ETA. Peaks: 1

SHORT

475

COMMUNICATIONS

assay precision produced a coefficient of variation of 4.91% (range 1.02% at 20 pg/ml to 5.79% at 0.2 pg/ml). _ The following drugs did not coelute with either ETA or PTA: acetaminophen, aspirin, amikacin, capreomycin, ethambutol, isoniazid, kanamycin, p-aminosalicyclic acid, pyrazinamide, streptomycin, and rifampin. Phenobarbital was shown to coelute with ETA. Concomitant treatment with phenobarbital should be avoided when measuring ETA serum concentrations. CONCLUSION

The HPLC method described for the determination of ethionamide in serum is specific for ETA, unlike bioassays which measure ETA and the active sulfoxide metabolite. The method shows sensitivity, specificity, and recovery similar to that described by Jenner and Ellard [3] and by Shepard et al. [5], while replacing the liquid-liquid extraction with a simple, reproducible solid-phase extraction method. REFERENCES 1 A. Kucers

and N. Mck. Bennett

(Editors),

The Use of Antibiotics, Lippencott,

Philadelphia,

PA, 4th

ed., 1988, pp. 14261431. 2 L. Heifets, Antimicrob. Agents Chemother., 32 (1988) 1131-l 136. 3 P. J. Jenner

and G. A. Ellard,

J. Chromatogr.,

225 (1981) 245251.

4 M. R. Holdiness, J. Chromatogr., 340 (1985) 321-359. 5 C. C. Shepard, P. J. Jenner, G. A. Ellard and R. D. Lancaster, 6 P. J. Jenner,

in D. S. Reeves and V. Villman

Medical Microbiology, Fisher Verlag,

Stuttgart,

(Editors),

Inc. J. Leprostl. 531 (1985) 587-594.

High Performance Liquid Chromatography in

New York,

1986, p. 153-165.